epiextractr

The epiextractr package makes it easy to use the EPI microdata extracts in R.

First download the CPS microdata using download_cps() into a directory on your machine. The downloaded data will simply be .feather files for each year of data.

For example, to download the Outgoing Rotation Group extracts into a directory called C:/data/cps, try

library(epiextractr)
download_cps("org", "C:/data/cps")

Then you can use load_org() to call a selection of years and columns:

load_org(2019:2021, year, female, wage, orgwgt, .extracts_dir = "C:/data/cps")

#> 
#> # A tibble: 824,963 × 4
#>     year orgwgt female      wage
#>    <int>  <dbl> <int+lbl>  <dbl>
#>  1  2019 11367. 1 [Female] 14   
#>  2  2019  6541. 1 [Female] 20.9 
#>  3  2019  6327. 0 [Male]    7.65
#>  4  2019  6327. 0 [Male]    7.65
#>  5  2019 11262. 1 [Female] 10   
#>  6  2019  7867. 1 [Female] 28.8 
#>  7  2019 11262. 1 [Female] 11   
#>  8  2019  7943. 0 [Male]   NA   
#>  9  2019  6092. 1 [Female] NA   
#> 10  2019  7738. 0 [Male]   NA   
#> # … with 824,953 more rows

For ease of use you can omit the .extracts_dir option by setting the following environment variables equal to the directory of the extracts:

EPIEXTRACTS_CPSBASIC_DIR
EPIEXTRACTS_CPSORG_DIR
EPIEXTRACTS_CPSMAY_DIR

For example, if your .Renviron file sets

EPIEXTRACTS_CPSORG_DIR=C:/data/cps

then you can simply run

load_org(2019:2021, year, female, wage, orgwgt)

More examples

Calculating employment rates by year and race

Calculate annual employment-to-population ratios by race/ethnicity from the 2010-2019 Basic CPS using tidyverse functions:

library(tidyverse)
load_cps("basic", 2010:2019, year, basicwgt, wbhao, emp) %>%
  filter(basicwgt > 0) %>%
  group_by(year, wbhao) %>%
  summarize(value = weighted.mean(emp, w = basicwgt)) %>%
  mutate(name = as.character(haven::as_factor(wbhao))) %>%
  pivot_wider(year)

#> 
#> # A tibble: 10 × 6
#> # Groups:   year [10]
#>     year White Black Hispanic Asian Other
#>    <int> <dbl> <dbl>    <dbl> <dbl> <dbl>
#>  1  2010 0.595 0.523    0.590 0.601 0.501
#>  2  2011 0.595 0.516    0.589 0.602 0.504
#>  3  2012 0.594 0.527    0.595 0.604 0.521
#>  4  2013 0.592 0.529    0.600 0.611 0.499
#>  5  2014 0.594 0.540    0.612 0.607 0.519
#>  6  2015 0.596 0.554    0.616 0.606 0.517
#>  7  2016 0.598 0.562    0.620 0.612 0.525
#>  8  2017 0.599 0.575    0.627 0.619 0.540
#>  9  2018 0.601 0.583    0.632 0.619 0.538
#> 10  2019 0.603 0.588    0.639 0.625 0.547